Video denoising method and system

Abstract

The invention discloses a video denoising method and belongs to the technical field of video processing. The video denoising method comprises performing separation of three color channels of R (red), G (green) and B (blue) on an source video; leading every color channel to be subjected to hard threshold grouping, collaborative filtering and aggregating in a four-dimensional space to obtain a hard threshold estimation video; leading the hard threshold estimation video to be subjected to Wiener filtering grouping, collaborative filtering and aggregating in the four-dimensional space to obtain denoised videos; and aggregating the denoised videos of the three color channels into an RGB three-dimensional video. According to the video denoising method, spatiotemporal correlation of spatiotemporal blocks of a video sequence is fully taken advantages of, spatial domain linear transformation and filtering transformation are performed in the four-dimensional space, noises are eliminated, meanwhile, fine structures in the video is well kept, and the spatiotemporal redundancy of the video sequence can be fully eliminated.

Description

technical field [0001] The invention relates to video processing technology, in particular to a video noise reduction method and system. Background technique [0002] With the emergence of a large number of digital images, surveillance videos, and various types of movies, especially for those images and video data acquired under unfavorable environments, there must be noise that affects image quality, so people are looking for high-quality image restoration methods. The need is more and more urgent, and video noise reduction has now become a widely researched topic in the field of computer vision. [0003] Video noise reduction is based on image noise reduction methods, and traditional image noise reduction methods can be divided into spatial domain noise reduction processing and transform domain filtering processing. The principle of spatial domain noise reduction is to use the correlation between adjacent pixels in the image, and the noise is independent and irrelevant to...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome that current video noise reduction methods do not make full use of the self-similar information of image regions, the technical problem to be solved by the present invention is to provide a video noise reduction method and system to make full use of the spatio-temporal correlation of spatio-temporal blocks in video sequences, and eliminate Preserve the fine structure in the video while maintaining the noise, fully eliminate the spatio-temporal redundancy of the video sequence, and improve the video quality

Images